This health services research study uses an existing NIMH-funded, epidemiological data set to examine whether homeless adults with alcohol problems are more likely to obtain work if they have societal contacts (i.e., social support from family and friends, social services, and treatment services). This timely study is informed by important changes in welfare reform policies that mandate work in conjunction with entitlement benefits. Homeless adults, many of whom have alcohol problems, obtain income from informal sector activities, public welfare entitlements and work. However, with the changes initiated by welfare reform, entitlement benefit income may be time-limited or may have a work requirement. Therefore, individuals who may have never held a full-time job will be seeking employment. Although current studies indicate that the vast majority of homeless adults work, at least sporadically, very few studies have examined factors associated with work. The proposed study will conduct secondary analyses using an existing longitudinal data set representing a county-wide, probability sample of homeless adults (N=564, 385 men and 179 women). Several statistical techniques including logistic regression and path regression models will be used to: (1) describe the nature and patterns of alcohol problems (and comorbidities) among homeless adults, and test the relationship between alcohol problems and work in this group; (2) document patterns of societal contact (such as family, friends, treatment, health care and social service agencies) and test its association with work among homeless adults; and then devise a composite scale representing societal contacts (i.e., contacts from family, friends, treatment, and social service agencies) among homeless adults; and (3) test the model that alcohol problems and societal contacts have an association with work.